Multifunctional T Lymphocytes Generated After Therapy With an Antitumor Gallotanin-Rich Normalized Fraction Are Related to Primary Tumor Size Reduction in a Breast Cancer Model

Abstract

Natural compounds are promising sources for anticancer therapies because of their multifunctional activity and low toxicity. Although the host immune response (IR) is clearly implicated in tumor control, the relationship between natural therapies and IR has not yet been elucidated. The present work evaluates IR induction after treatment with a gallotannin-rich fraction from Caesalpinia spinosa (P2Et). Breast tumor 4T1 cells were used to evaluate antitumor properties and IR activation. Apoptosis and expression of immunogenic cell death (ICD) markers were assessed in vitro, whereas IR and postvaccination tumor evolution were assessed in vivo. P2Et fraction induced apoptotic cell death, displaying phosphatidylserine externalization and DNA fragmentation. ICD markers such as calreticulin, high-mobility group box 1 translocation from nuclei to cytoplasm, and ATP secretion were observed. Primary tumor control was improved by vaccination with P2Et-pretreated 4T1 cells (t-P2Et), yielding long-lasting ex vivo multifunctional CD4⁺ and CD8⁺ T lymphocytes (interleukin [IL]-2⁺, tumor necrosis factor [TNF]-α⁺, interferon [IFN]-γ⁺) that secrete IL-2, TNF-α, IL-4, IL-5, and IFN-γ after specific 4T1 cell stimulation. The present study constitutes the first demonstration of a long-lasting antitumor IR induction and primary tumor reduction induced by a complex natural fraction. These data reveal the potential use of this fraction as an adjuvant in breast cancer treatment.

Background

Although the worldwide mortality rate in breast cancer has decreased since 2009, breast cancer continues to have the highest incidence and is the second leading cause of cancer death in women in the United States. ¹ Current therapies are effective, but resistance develops after prolonged use. Immune response (IR) has been implicated in the effectiveness of therapies after evaluating pathologically complete responses as well as the survival of patients with various types of breast cancer treated with neoadjuvant chemotherapy ² and other tumors (after surgical resection of the primary tumor). ³ This connection confirms the importance of the IR as a positive prognostic factor, particularly after treatment with anthracyclines and oxaliplatin. ⁴ In addition, immunocompromised individuals do not respond to chemotherapy. ⁵

Adaptive antitumor IRs occur after the activation of T lymphocytes (TLs) by antigen presenting cells (APCs) that were previously activated by danger signals secreted during tumor cell death. ⁶ The mechanisms through which APCs effectively induce the TL response in cancer have been grouped under the concept of immunogenic cell death (ICD). This phenomenon has been described in tumor cells undergoing death in response to anthracyclines, oxaliplatin, cyclophosphamides, or irradiation and results in the induction of specific CD4⁺ and CD8⁺ TLs. ⁷ Only cells with calreticulin (CRT) at the plasma membrane, which release ATP (adenosine triphosphate) ⁸ or high-mobility group box (HMGB1) ⁹ are able to induce an IR.^10,11 The protective IR is generated by a potent activation signal during priming to generate long-lived memory TLs. ¹²

Plants are an important source of medicines, and the earliest records of approximately 1000 plant-derived substances date from approximately 2600 bce in Mesopotamia. ¹³ The use of plants from traditional Chinese medicine (TCM) with conventional chemotherapy is an effective technique to significantly increase the survival of patients with breast cancer, ¹⁴ hepatocellular carcinoma, ¹⁵ lung carcinoma, ¹⁶ and colon cancer. ¹⁷ Components of both innate and adaptive immunity are potentially modulated by specific TCMs. ¹⁰ In fact, some evidence suggests that plant extracts used in folk medicine induce an antitumor IR. For example, the Japanese traditional medicine Juzen-taiho-to has been implicated in the induction of antitumor IRs in the Mel-RET melanoma model. ¹⁸ Additionally, some natural compounds induce endoplasmic reticulum stress, which can be related to the secretion of danger signals ¹⁹ or to autophagy.

Recently, we obtained a gallotannin-rich fraction from Caesalpinia spinosa (P2Et), a shrub commonly named dividivi, which shows antimicrobial, ²⁰ antioxidant, and antitumor activity. P2Et contains hydrolysable tannins, such as galloylquinic acid derivatives, in high proportions and pentagalloylglucose and gallic acid–containing compounds (gallates) in lower proportions. It induces apoptosis in several tumor cells and sensitizes K562 (chronic myelogenous leukemia) cells to doxorubicin (DX)-induced death. ²¹ Treatment with the P2Et fraction improves primary tumor control and decreases the number of metastatic cells in the spleen and lung in the highly invasive and poorly immunogenic murine breast adenocarcinoma cell line, 4T1; these decreases are concomitant with a reduction of interleukin (IL)-6 serum levels. ²² In the present study, we provide evidence demonstrating that a specific IR is generated by this natural fraction, preceded by HMGB-1 mobilization and CRT expression on the plasma membrane of 4T1 tumor cells. These results suggest that ICD might the basis of IR generation and reveal a possible explanation for the biological activity of natural products, including the fraction derived from C spinosa.

Materials and Methods

Plant Material, Extraction, and Purification

C spinosa pods were collected in Villa de Leyva, Boyacá, Colombia, in March 2007 and identified by Luis Carlos Jiménez from the Colombian National Herbarium, voucher specimen number COL 523714 (Colombian Environmental Ministry agreement number 0454 related to the use of genetic resources and derived products). The P2Et fraction was purified and characterized as previously described. ²¹

Reverse Phase- High Performance Liquid cromatography (HPLC) Conditions

Chromatographic fingerprinting was performed by injecting 5 µL of C spinosa fraction (1 mg/mL) and was analyzed on a HPLC Agilent 1200 series with an Agilent 1200 Photodiode array (PDA) detector (λ = 254 nm; Agilent, Wilmington, DE). Compounds were separated on Agilent Eclipse XD8 C18 Column (4.6 × 250 mm, 5 µm): mobile phase A, acetonitrile; mobile phase B, H₃PO₄ 0.1%; gradient time: 0 minutes, 5%B; 10 minutes, 12%B; 30 minutes, 21%B; 60 minutes, 30%B; postrun: 5 minutes. Acquisition and data analysis were performed using Chemstation v 3.0 software. Gallic acid from Sigma was used as a control compound and was quantified externally by a standard method.

Tumor Cell Lines and Culture Conditions

The 4T1 mammary carcinoma is a transplantable tumor cell line. The tumor grows in BALB/c mice and in tissue culture. 4T1 was selected without mutagenesis for its resistance to 6-thioguanine. The 4T1 tumor is highly tumorigenic and invasive and, unlike most tumor models, can spontaneously metastasize from the primary tumor in the mammary gland to multiple distant sites. ²³ The murine mammary tumor cell line, 4T1, was provided by Dr Alexzander Asea (Texas A&M Health Science Center College of Medicine, Temple, TX). The B16 murine melanoma is a nonlymphoid solid tumor, which originated spontaneously in the ear of a female C57BL/6J. The murine melanoma tumor cell line B16 was provided by Dr Pedro Romero (Ludwig Center for Cancer Research, Department of Oncology, Faculty of Biology and Medicine University of Lausanne, Switzerland). Cells were cultured in RPMI-1640 (Eurobio, Toulouse, France) supplemented with heat-inactivated fetal calf serum (10%; Eurobio), 2 mM l-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, 0.01 M Hepes buffer, and 1 mM sodium pyruvate (Eurobio) and incubated in a humidified environment at 37°C and 5% CO₂. The cells were selected by culturing with 6-thioguanine, and metastatic ability was tested by detecting tumor cells in lung and spleen. Tumor cells were proven Mycoplasma free using a MycoProbe Mycoplasma Detection Kit (R&D Systems) and maintained with ciprofloxacin (0.5 µg/mL). Cells were grown to 75% confluency and subcultured using trypsin/EDTA 1X (Eurobio), washed with phosphate-buffered saline (PBS), and resuspended in supplemented RPMI-1640.

Quantification of Cell Death

Cell death was quantified using Annexin V-Alexa Fluor 488 (Molecular Probes, Invitrogen Corp, Carlsbad, CA, USA) and propidium iodide (PI; Sigma-Aldrich, St Louis, MO, USA), as previously reported. ²¹ Triplicate samples were acquired on a FACSAria II (Becton Dickinson, San Jose, CA, USA) and analyzed with FlowJo software (Tree Star Inc, Ashland, OR, USA).

Immunofluorescence Analysis

For immunofluorescence analyses, 4T1 cells (3 × 10⁵) were treated with the P2Et fraction (34 µg/mL), DX (positive control, 0.51 µg/mL), or ethanol (negative control, 0.02%). Nuclei and actin were stained with DAPI (4′,6-diamidino-2-phenylindole; Sigma-Aldrich) and Alexa Fluor 594 phalloidin (Molecular Probes, Invitrogen), respectively, as previously reported. ²⁴ HMGB-1 detection was performed as previously reported. ²⁵ Briefly, treated cells (4T1 and B16F10) were stained with a primary antibody (rabbit polyclonal antibody against HMGB1, Abcam) and then for 30 minutes with an Alexa Fluor 488-conjugated secondary antibody (Molecular Probes, Invitrogen), followed by DAPI for 5 minutes. In both cases, slides were mounted using a ProLong Antifade Kit (Molecular Probes). The cells were imaged with a laser scanning confocal microscope FV1000 (Olympus) using an UPLSAPO 60X 1.35 NA oil immersion objective. A 405-nm diode laser was used to visualize DAPI; a 488-nm argon laser line was used for Alexa Fluor 488 nm; and a HeNe 543-nm laser was used for Alexa Fluor 594 nm. XY sections at 0.207 or 0.165 µm per pixel and a resolution of 1024 × 1024 or 640 × 640 were obtained.

Animals

Female BALB/c mice (6 to 12 weeks old) were purchased from Charles Rivers Laboratories International, Inc, Boston, MA, USA and housed in our animal research facility following the established protocols of the Ethics Committee of the Science Faculty and National and International Legislation for Live Animal Experimentation (Colombia Republic, Resolution 08430, 1993; National Academy of Sciences, 2010). Mice were housed in polyethylene cages with food and water ad libitum, controlled temperature, and a 12-hour light/dark cycle. Before treatment, mice were acclimated for 1 week under standard conditions. This project was approved by the ethics committee of the Science Faculty on August 21, 2007.

Vaccination and Tumor Transplantation

To evaluate the effect of P2Et treatment on the IR, we used 2 models. In the first model, 4T1 cells (3 × 10⁶) treated with the P2Et fraction (34.1 µg/mL; t-P2Et) or ethanol (negative control, 0.02%; t-live) for 48 hours were injected into the footpad of BALB/c mice. After 1 week, the popliteal lymph nodes (PLNs) were harvested and stained, then observed using flow cytometry analysis. In the second model, 4T1 cells (3 × 10⁶) treated with the P2Et fraction (34.1 µg/mL; t-P2Et) or ethanol (negative control, 0.02%; t-live) for 48 hours were injected into the footpad of BALB/c mice on day −6. Then, 4T1 cells (1 × 10⁴) were suspended in PBS (100 µL) and injected into the right mammary fat pad (subcutaneously) on day 0. After 6 days, vaccinated mice were restimulated with t-P2Et (3 × 10⁶) or t-live (3 × 10⁶) in the footpad or were treated intraperitoneally with the P2Et fraction (18.7 mg/kg). Mice vaccinated with t-P2Et in the footpad or treated intraperitoneally with P2Et were dosed 3 times with the P2Et fraction (18.7 mg/kg). Tumors were measured with Vernier calipers twice a week, and the mice were killed humanely 21 days post–tumor implantation. The tumor volume was calculated using the following formula: tumor volume (mm³) = [(Width)² × Length]/2.

Flow Cytometry Analysis

To evaluate CRT expression, a total of 2 × 10⁵ cells were plated in 12-well plates and treated the following day with P2Et, DX (positive control), or ethanol at the same concentration for 24 hours. Cells were stained as previously reported. ²⁵ The cells were incubated for 30 minutes with the primary antibody (rabbit polyclonal antibody against calreticulin, Abcam, Cambridge, UK) followed by the Alexa488-conjugated monoclonal secondary antibody in blocking buffer for an additional 30 minutes (Molecular Probes, Invitrogen). Each sample was then processed using a FACSAria II (Becton- Dickinson) and analyzed with FlowJo software (Tree Star Inc) to identify the cell surface CRT. The fluorescent intensity of stained cells was assessed within the aqua-negative cell gate.

For the priming assay, PLNs from vaccinated mice were harvested as previously described, and cells were surface stained with anti-CD3-Alexa Fluor 647, anti-CD4-PerCP, or CD8-PE (BD Pharmingen). Then, for intracellular IL-2, tumor necrosis factor (TNF)-α, and interferon (IFN)-γ expression, cells were restimulated or not with 4T1 cell lysate (performed using 3 × 10⁵ tumor cells killed by 5 minutes heating at 42°C followed by 1 cycle of freezing/thawing in liquid nitrogen) for 72 hours and were then permeabilized with Cytofix/Cytoperm (BD Biosciences), washed twice with Perm/Wash TM (BD Biosciences), and stained with anti-IL-2-FITC, anti-TNF-α-PE-Cy7, and anti-IFN-γ-Alexa Fluor 700 (all from BD Pharmingen) for 30 minutes at 4°C. The supernatant was harvested 72 hours later, and IFN-γ, IL-4, and IL-5 secretion was assessed by ELISA (IFN-γ) and CBA (IL-4 and IL-5; from BD Pharmingen). The cells were then harvested and surface stained with antibodies specific for CD3, CD4, or CD8, fixed and permeabilized (Cytofix/Cytoperm kit, BD), and then labeled with anti-IL-2, anti-TNF-α, and anti-IFN-γ antibodies. Alternatively, the cells were stimulated with Phorbol 12-myristate 13-acetate (PMA) and ionomycin (P/I) or 4T1 lysate for 1 hour; followed by the addition of brefeldin A (6 hours at 37°C), surface stained with antibodies specific for CD3, CD4, or CD8; and after fixation and permeabilization (Cytofix/Cytoperm kit), labeled with anti- IFN-γ, IL-2, and TNF-α antibodies. Immunofluorescence was analyzed with a FACSAria II (BD Biosciences), with FACS Diva software and with FlowJo software ((Tree Star Inc).

ATP Assays

To evaluate ATP release, a total of 1 × 10⁶ cells were treated the following day with P2Et (34 µg/mL), DX (positive control, 0.51 µg/mL), or ethanol for 15 and 24 hours, and extracellular ATP was measured by ATP Bioluminescence Assay Kit HS II (Roche, Germany) following the manufacturer’s instructions. For assessment of the chemoluminescent signal, the plates were read in a Chameleon V (FI-TRF-FP-Abs-Lum reader; Hidex).

Statistical Analysis

Statistical analysis was performed with GraphPad Prism software version 5.0 (Graph Pad Prism Software Inc, San Diego, CA) using the nonparametric Mann-Whitney U test. Significance was set at P < .05. Data are shown as the medians unless otherwise stated. For multiparametric analyses, the SPICE 5.2 Software was used, and Boolean gating analyses were performed.

Results

Chromatographic Analysis

C spinosa fraction was obtained as previously described and used to standardize chromatographic conditions. Best resolution was obtained using Acetonitrile/H₃PO₄ gradient. The chromatogram (Figure 1) reveals almost 70 peaks. Gallic acid (retention time [RT] = 5.724 minutes) and ethyl gallate (RT = 22.947 minutes) were isolated from the total fraction, and its structure was confirmed by mass spectrometry and nuclear magnetic resonance (data not shown); RT was confirmed by comparison with commercial standards.

Figure 1.

Chromatographic analysis of gallotannin-rich fraction from Caesalpinia spinosa: gallic acid (retention time [RT] = 5.724 minutes; Molecular weight [MW]: 169) and ethyl gallate (RT = 22.947 minutes; MW: 197).

P2Et Treatment of Murine 4T1 Breast Cancer Cells Induces ICD

We provide evidence that the P2Et fraction induces apoptosis in 4T1 murine breast cancer cells, as observed by phosphatidylserine exposure and DNA fragmentation (Figures 2A and 2B), confirming our previous results. ²² In addition, the P2Et fraction induces CRT expression on 4T1 cells at a lesser but comparable level to DX (Figure 2C). Additionally, P2Et induced HMGB-1 mobilization from nuclei into the cytoplasm, similar to DX treatment (Figure 2D) and ATP release (Figure 2E), suggesting that classical apoptosis cell death followed by the expression of ICD markers potentially promotes the activation of a specific IR. Similar results were observed with the murine melanoma cell line, B16 (Supplementary Figure 1).

Figure 2.

Gallotannin-rich fraction from Caesalpinia spinosa (P2Et) fraction induces immunogenic cell death in mammary cancer cell. A. Frequency of apoptotic (Annexin V+, propidium iodide [PI]−) and secondary necrotic (Annexin V+, PI+) 4T1 cells after 24 and 48 hours of treatment with the P2Et fraction, doxorubicin (DX), or ethanol (as negative control). 4T1 cells were stained with Annexin V-Alexa Fluor 488 and with the viability dye PI before subsequent analysis by flow cytometry. Data are presented as the mean value ± standard error of the mean (SEM) of 1 of 5 experiments. B. 4T1 cells were treated with the P2Et fraction, DX, or vehicle control for 48 hours and stained with DAPI (4′,6-diamidino-2-phenylindole) and Alexa Fluor 594-phalloidin. White arrows show apoptotic cells or fragmented nuclei observed by confocal microscopic examination. Calreticulin (CRT) exposure on the cell surface of 4T1 cells was assessed after stimulation with the P2Et fraction, DX, or ethanol (as negative control) by flow cytometry. C. Representative histogram, frequency of CRT cells, and Mean fluorescence intensity [MFI] of CRT cells. Data are presented as the mean ± SEM of 5 independent experiments. D. Confocal microscopy of high-mobility group box (HMGB1) released from the nuclei to the cytosol in 4T1 cells treated with the P2Et fraction, DX, or ethanol for 24 hours. After the indicated treatments, cells were stained with antirabbit HMGB1 antibody and with a goat antirabbit Alexa 488. Nuclei were stained with DAPI. E. Extracellular ATP (adenosine triphosphate) after P2Et, DX, or ethanol treatment for 15 and 24 hours. Data are presented as the mean ± SEM of 3 independent experiments.

Vaccination With t-P2Et Stimulates the Early Recruitment of PLN CD3⁺ TL and Favors Early Generation of CD4⁺ and CD8⁺ TLs That Spontaneously Produce IL-2 and TNF-α

Because the treatment of tumor cells with the P2Et fraction induced some signals associated with ICD, we evaluated whether the vaccination of BALB/c mice with t-P2Et could generate an antitumor IR, as has been demonstrated in other murine cancer models, including colon carcinoma (CT26).^25,26 The vaccination of mice with t-P2Et promotes the recruitment of mononuclear cells, mainly CD3⁺ (Supplementary Figures 2A, 2B, and 6), with a preferential recruitment of CD4⁺/IL-2⁺/ TNF-α⁺ TL to PLNs (Figures 3A and 3B, Supplementary Figure 3). These cells did not generally secrete IFN-γ (Figure 3C). In contrast, CD4⁺/ IFN-γ⁺ TLs were significantly increased after P/I stimulation in t-P2Et-vaccinated mice (Supplementary Figure 3), suggesting an enhanced sensitization process after t-P2Et vaccination but which is insufficient to induce the maturation of IFN-γ⁺ effector cells in the absence of secondary stimuli. Vaccination with t-P2et also favored the generation of CD8⁺/IL-2⁺/TNF-α⁺ (Figures 3D and 3E). However, as observed in CD4⁺ TLs, the CD8⁺/IFN-γ⁺-producing TLs were principally detected after in vitro P/I restimulation (Supplementary Figure 3). In summary, we observed that P2Et vaccination induces the early recruitment of LT CD3⁺, which are actively secreting cytokines and are capable of generating effector cells on in vitro restimulation.

Figure 3.

Vaccination with 4T1 cells (3 × 10⁶) treated with the gallotannin-rich fraction from Caesalpinia spinosa (t-P2Et) favors early generation of CD4⁺ and CD8⁺ T lymphocytes (TLs) that spontaneously produce interleukin (IL)-2 and tumor necrosis factor (TNF)-α and the generation of multifunctional CD4⁺ TLs in mice. 4T1 cells were treated with the P2Et fraction, doxorubicin (DX), or ethanol (negative control, t-live) for 48 hours. After treatment, 3 × 10⁶ cells of each treatment were injected into the footpad of BALB/c mice; 7 days later, the Peripheral Lymph nodes [PLNs] of each group (phosphate-buffered saline [PBS], t-live, t-P2Et, and t-DX) were obtained. The CD3⁺/CD4⁺ TLs producing (A) IL-2, (B) TNF-α, and (C) interferon (IFN)-γ as well as CD3⁺/CD8⁺ TLs producing (D) IL-2, (E) TNF-α, and (F) IFN-γ were determined by flow cytometry. Data represent the median values and ranges of 3 independent experiments with 7 or 8 mice per group. (***P < .0001, **P < .001, *P < .01). (G) The cells were restimulated with 4T1 cell lysate or PBS for 1 hour, 72 hours after treatment; frequency of CD4⁺ TL producing TNF-α by flow cytometry. (H) Frequency of CD3⁺/CD4⁺ or CD3⁺/CD8⁺ TLs producing 1, 2, or 3 cytokines, such as IL-2, TNF-α, and IFN-γ by flow cytometry. The pie charts were generated using SPICE software. Data are presented as median values and ranges of 3 independent experiments with 7 or 8 mice per group. (***P < .0001, **P < .001, *P < .01).

Vaccination With t-P2Et Induces Multifunctional, Antigen-Specific TLs More Effectively Than t-DX

To determine the quality and specificity of the TL response induced in our model and hence the occurrence of a cross-priming mechanism, Th1 profiles of CD4⁺ and CD8⁺ TLs obtained from BALB/c mice vaccinated with t-P2Et, t-DX, live 4T1, or PBS only were assessed. PLNs obtained after 7 days of vaccination were cultured for 72 hours in the presence of anti-CD28, anti-CD49d, and 4T1 lysate or PBS to give a specific antigenic signal. The frequency of CD4⁺ and CD8⁺ TLs producing IL-2, TNF-α, and IFN-γ was determined, and soluble cytokines were analyzed. We observed that vaccination with t-P2Et favored the generation of a CD4⁺/TNF-α⁺ population (Figure 3G) preferentially responding to the 4T1 lysate. Using a Boolean analysis to evaluate multifunctionality, we identified the frequency of CD4⁺ and CD8⁺ TLs producing 1, 2, or 3 different cytokines. As shown in Figure 3H, t-P2Et induces a higher frequency of specific multifunctional CD4⁺ TLs producing IFN-γ⁺/IL-2⁺/TNF-α⁺ (0.0300% ± 0.0006%) when compared with the PBS group (0.0020% ± 0.002%) or with the t-live treated mice (0.0006% ± 0.004%). This translates into a greater ability of t-P2Et to generate a quick and active effector T cell response. In contrast, the treatment with t-DX only favors the generation of CD4⁺ TLs producing more IL-2⁺/TNF-α⁺ (0.060% ± 0.014%) than PBS (0.005% ± 0.004%) or t-live (0.020% ± 0.008%; Figure 3H). These results suggest that the treatment of tumor cells with P2Et favors the priming of TLs able to recognize tumor cells after a second Ag encounter. For CD8⁺ TLs, we observed that treatment with t-P2Et favors a small increase in the frequency of multifunctional CD8⁺ TLs (0.0300% ± 0.002%) when compared with PBS (0.0020% ± 0.001%) or t-live (0.0006% ± 0.003%). Treatment with t-DX, unlike treatment with t-P2Et, did not generate a significant percentage of multifunctional TLs or IFN-γ⁺/TNF-α⁺ TLs (Figure 3H). In summary, gallotannin-rich P2Et therapy induces the generation of multifunctional, antigen-specific TL more effectively than DX.

Vaccination With t-P2Et Retards Primary Tumor Growth and Favors the Permanence of a Multifunctional T Cell Population in PLNs

The induction of an effector response secondary to vaccination involves an effective priming process as well as the presence of survival signals that allow for a TL response to a second stimulus with antigen. Given that the P2Et fraction induces the secretion of danger signals by tumor cells, ²² we hypothesized that vaccinating BALB/c mice with t-P2Et might have an effect on tumor growth and potentiate the effect of drug treatment. Indeed, the role of vaccination in the control of mouse tumors has been observed in other murine cancer models. ²⁶ As observed in Figure 4A, the immunization of BALB/c mice with t-P2Et followed by P2Et treatment twice a week significantly delays primary tumor growth when compared with mice treated intraperitoneally with P2Et but without vaccination. These results clearly implicate the IR in the control of early implantation and tumor development.

Figure 4.

Vaccination with t-gallotannin-rich fraction from Caesalpinia spinosa (P2Et) retards primary tumor growth. 4T1 cells were treated with the P2Et fraction for 48 hours (t-P2Et). On day −6, animals were randomly assigned to the control or treatment groups. The mice were inoculated with t-P2Et or phosphate-buffered saline (PBS) into the footpad. On day 0, all groups of mice were inoculated with 4T1 live cells (subcutaneously). On day 6, the mice were treated again as described on day −6 or with the P2Et fraction (18.7 mg/kg) via intraperitoneal injection. On day 20, mice were killed humanely, and the popliteal lymph nodes (PLNs) were obtained. A. Growth of the primary tumor. B. PLN cell number. C. CD3⁺ and CD3⁻ T lymphocytes PLN frequency. Data are expressed as the mean ± standard error of the mean or median and range of 2 independent experiments, with 7 to 8 mice per group (***P < .0001, **P < .001, *P < .01).

Further to our earlier observation that t-P2Et vaccination induces the recruitment of CD3⁺ TLs in PLNs as early as 7 days postvaccination, mice vaccinated with t-P2Et continued to have a significant number of cells in the PLNs after 20 days of treatment compared with the control mice (Figure 4B). However, at this time point, cells were principally CD3⁻ (Figure 4C). Studies are now under way in our laboratory to evaluate the cell migration kinetics of different mononuclear cells to the primary tumor or metastatic sites after vaccination or treatment and the role of each cell type on the tumor control.

Vaccination under optimal conditions should generate a specific, long-lasting CD4⁺ and CD8⁺ TL response, easily detectable in the PLNs adjacent to the tumor area. In our model (Figure 5A), we showed that vaccination inhibited tumor growth after 20 days of evaluation and resulted in detectable CD4⁺/IL-2⁺ TLs ex vivo (Figure 5B) and after restimulation with P/I; this result is not observed for mice treated with PBS (Supplementary Figure 4). CD4⁺/TNF-α⁺ TLs also significantly increase ex vivo (Figure 5B) but, in contrast, CD4⁺/IFN-γ⁺ TL was only evident after P/I stimulation (Figure 5B and Supplementary Figure 4). Moreover, we detected only a slight increase in the number of CD8⁺/IL-2⁺ cells ex vivo (Figure 5B), with no changes in TNF-α-producing cells (Figure 5B). Conversely, we observed an increased number of IFN-γ-producing cells both ex vivo and after P/I stimulation, suggesting that the generation of a robust effector response (Figure 5B and Supplementary Figure 4F) may be implicated in tumor growth control.

Figure 5.

Mice vaccinated with 4T1 cells (3 × 10⁶) treated with the gallotannin-rich fraction from Caesalpinia spinosa (t-P2Et) retain a population of activated T cells in popliteal lymph nodes (PLNs). 4T1 cells were treated with the P2Et fraction for 48 hours (t-P2Et). On day −6, animals were randomly assigned to the control or treatment groups. The mice were inoculated with t-P2Et or phosphate-buffered saline (PBS) into the footpad. On day 0, all groups of mice were inoculated with 4T1 live cells (subcutaneously). On day 6, the mice were treated again as described on day −6 or with the P2Et fraction (18.7 mg/kg) via intraperitoneal injection. On day 20, mice were sacrificed, and the PLN were obtained. A. Schematic representation of the vaccination. B. The numbers of CD3⁺/CD4⁺ T lymphocytes (TLs) producing interleukin (IL)-2, tumor necrosis factor (TNF)-α, and interferon (IFN)-γ as well as CD3⁺/CD8⁺ TLs producing IL-2, TNF-α, and IFN-γ were determined by flow cytometry. Data are expressed as the mean ± standard error of the mean or median and range of 2 independent experiments with 7 to 8 mice per group (***P < .0001, **P < .001, *P < .01).

Additionally, we also evaluated cytokine production and IR after cells were cultured with anti-CD28, anti-CD49d, and 4T1 lysate after 20 days of vaccination to evaluate if the early-detected response was a long-lasting response. Vaccination with t-P2Et favors the permanence of CD4⁺/IL-2⁺ TLs without stimulation and after 4T1 lysate culture, suggesting the presence of a preactivated subset of T cells (Figure 6A). These results indicate that CD4⁺ TLs received a fitness signal during priming to promote survival in vivo after a second Ag stimulation (tumor transplanted at day 0). Regarding the generation of CD4⁺/TNF-α⁺, CD4⁺/IFN-γ⁺, and CD8⁺/IFN-γ⁺, no statistically significant differences were observed with the different treatments (Supplementary Figure 5). Similar results were observed for CD8⁺/IL-2⁺ and CD8⁺/TNF-α⁺ cells (Figures 6B and 6C). It is possible that effector cell homing to the tumor occurs after the priming. These results strongly suggest that vaccination with t-P2Et generates an IR in the lymph nodes adjacent to the tumor, which remains active 20 days after treatment.

Figure 6.

Generation of CD4⁺ and CD8⁺ T lymphocytes (TLs) producing interleukin (IL)-2. 4T1 cells were treated with the gallotannin-rich fraction from Caesalpinia spinosa (P2Et) fraction for 48 hours (t-P2Et). On day −6, animals were randomly assigned to the control or treatment groups. The mice were inoculated with t-P2Et or phosphate-buffered saline (PBS) into the footpad. On day 0, all groups of mice were inoculated with 4T1 live cells (subcutaneously). On day 6, the mice were treated again as described on day −6 or with the P2Et fraction (18.7 mg/kg) via intraperitoneal injection. On day 20, the mice were killed humanely, and the popliteal lymph nodes were obtained and cultured with anti-CD28, anti-CD49d, and 4T1 cell lysate or PBS. After 72 hours, the cells were restimulated with 4T1 cells lysate or PBS, and the numbers of CD3⁺/CD4⁺ TLs producing (A) IL-2 and CD3⁺/CD8⁺ TLs producing (B) IL-2 or (C) TNF-α were determined by flow cytometry. Data are expressed as the mean ± standard error of the mean or median and range of 2 independent experiments with 7 to 8 mice per group (***P < .0001, **P < .001, *P < .01).

IL-4-, IL-5-, and IFN-γ-Secreting Cells Specific of 4T1 Are Generated After t-P2Et Vaccination and Remain at the PLNs

The cytokines Th1 and Th2 have been implicated in tumor growth control. ²⁷ We observed that vaccination with t-P2Et or t-DX induced the generation and activation of IL-4, IL-5, and IFN-γ in T cells after in vitro stimulation with 4T1 cell lysate (Figures 7A, 7B, and 7C) compared with PBS. Interestingly, vaccination with t-P2Et significantly favors the early generation of T cells producing IL-5, whereas vaccination with t-DX mainly generated cells producing IFN-γ. These IL-4-, IL-5-, and IFN-γ-producing cells are also detected after 20 days of vaccination and tumor transplantation and are evident after stimulation with 4T1 cell lysate, suggesting the presence of memory TLs. In addition, spontaneous cytokine production is observed and may be a result of the presence of other cells, such as NK (Figures 7D, 7E, and 7F). Taken together, these results suggest that different priming and activation mechanisms are triggered after vaccination and may be related to death signals induced by P2Et in comparison to DX.

Figure 7.

Vaccination with 4T1 cells (3 × 10⁶) treated with the gallotannin-rich fraction from Caesalpinia spinosa (P2Et) promotes the generation and popliteal lymph node (PLN) recruitment of interleukin (IL)-4-, IL-5-, and interferon (IFN)-γ-secreting cells responding to specific Ag stimulation in mice. Upper panel: 4T1 cells were treated with the P2Et fraction (t-P2Et), doxorubicin (t-DX), or ethanol (t-live) for 48 hours. After treatment, 3 × 10⁶ cells from each treatment were injected into the footpad of BALB/c mice; 7 days later, PLNs of each group were obtained and cultured with anti-CD28, anti-CD49d, and 4T1 cell lysates or phosphate-buffered saline (PBS). Supernatants were harvested 72 hours later. (A) IL-4, (B) IL-5, and (C) IFN-γ secretion were assessed by ELISA (IFN-γ) and CBA (IL-4 and IL-5). Data are presented as median values and ranges of 3 independent experiments with 7 or 8 mice per group. Bottom panel: 4T1 cells were treated with the P2Et fraction for 48 hours (t-P2Et). On day −6, animals were randomly assigned to the control or treatment groups. The mice were inoculated with t-P2Et or PBS into the footpad. On day 0, all groups of mice were inoculated with 4T1 live cells (subcutaneously). On day 6, the mice were treated again as described on day −6 or with the P2Et fraction (18.7 mg/kg) via intraperitoneal injection. On day 20, mice were killed humanely, and the PLNs were obtained and cultured with anti-CD28, anti-CD49d, and 4T1 cell lysates or PBS. After 72 hours, the cells were restimulated with 4T1 cell lysate or PBS. The supernatant was harvested 72 hours later. (D) IL-4, (E) IL-5, and (F) IFN-γ were assessed by ELISA (IFN-γ) and CBA (IL-4 and IL-5). Data are expressed as the mean ± standard error of the mean or median and range of 2 independent experiments with 7 to 8 mice per group (***P < .0001, **P < .001, *P < .01).

Discussion

Breast cancer is the most common tumor in women worldwide, comprising an estimated 25% (1.67 million) of new cases in 2012 according to Globocan (http://globocan.iarc.fr/Default.aspx). In Colombia, breast cancer is the second most common cancer in women, with an incidence rate of 23.4% in 2012. According to the National Cancer Institute, 698 new cases of breast cancer, which corresponds to 11.4% of all new cancer cases (National Cancer Institute, 2011) were recorded in 2011. Metastasis is the leading cause of mortality in breast cancer. Although current therapies are effective against primary tumors, resistance develops after extended use. ²⁸

In normal practice, surgery and radiation therapy are the local treatments to reduce the risk of cancer in the breast, chest wall, and regional lymph nodes, whereas chemotherapy and hormonal therapy are the systemic treatments to reduce relapse and overall mortality.^29,30 Patients receiving radiation and chemotherapy treatment experience adverse effects, which are the major obstacle for successful treatment. Furthermore, the best-possible treatments are largely ineffective in advanced stages where metastasis has already occurred; therefore, there is an imperative need to develop chemopreventive, nontoxic agents, which are effective in treating metastatic disease. The IR has been implicated in the effectiveness of therapies after evaluating pathological complete response and survival of patients with different types of breast cancer treated with neoadjuvant chemotherapy ² as well as other tumors (after surgical resection of the primary tumor), ³ confirming the importance of the IR as a positive prognostic factor, particularly after treatment with anthracyclines and oxaliplatin.^4,31-33 In addition, immune-compromised individuals do not respond to chemotherapy.^5,34

In this regard, dietary botanicals have attracted considerable attention because of their intriguing biological activities at nontoxic levels. Plants are a major source of drugs, and the first reports of approximately 1000 substances derived from plants occurred in Mesopotamia in 2600 bce. ¹³ The use of plants is widespread in TCM; plant-derived substances in combination with conventional chemotherapy significantly increase the survival of patients with breast cancer and other cancers, thereby making plants valuable in the search for new cancer therapy candidates. ¹⁴ For example, observational studies have correlated green tea intake with a reduced risk of breast cancer incidence and recurrence.^{35 -37} The most abundant and possibly most potent polyphenol in green tea is epigallocatechin-3-gallate (EGCG), ³⁸ implied in the inhibition of hepatocyte growth factor, having as a consequence inhibition of tumor cell growth. Additionally, EGCG reduces migration and invasion through the inhibition of the angiogenic signal mediated by the vascular endothelial growth factor.^39-41

Regarding the immune system, inhibition of mitogen or antigen TL proliferation have been observed with 50 µg/mL (100 µM) of EGCG. ⁴² However, taking into account that apoptosis of tumor cells, or even fibroblasts, can be induced from 40 to 250 µM, it is possible that 100 µM induces TL cell death.^43,44 We do not measure the direct effect of the gallotanin-rich fraction on TL proliferation in vitro because it is not a good measure of adjuvant activity. In contrast, we observed that vaccination of mice with the t-P2Et fraction stimulates the early recruitment of PLN CD3⁺ TLs, producing IL-2 and TNF-α after a pulse with 4T1 cells, which demonstrates the capacity for activation and proliferation of these cells. In this way, Tae Heung Kang found that CD11c⁺ enriched cells isolated from mice treated with 0.5 mg/mL EGCG were more effective in stimulating E7-specific CD8⁺ T cells to secrete IFN-γ when compared with CD11c⁺ enriched cells from mice not treated with EGCG, ⁴⁵ suggesting good cross-priming processes. It is possible that ICD induced by t-P2Et leads to a strong specific T cell activation, as shown later in our model, where the presence of IFN-γ-, IL-4-, and IL-5-secreting cells responding to specific Ag stimulation was observed in mice.

We have previously shown that P2Et, a polyphenol-rich fraction with antitumoral activity, contains a large amount of gallic acid after acid hydrolysis, mainly because of the presence of gallic acid derivatives like galloylquinic acids, which has been reported for pods by Clifford et al, ⁴⁶ and our group. We optimize a chromatographic fingerprint using HPLC/PDA, and results revealed that most compounds in the fraction absorb at a wavelength of 254 nm, characteristic of some polyphenol compounds and gallotannins. In addition, we identified free gallic acid (molecular weight = 169) at 5.724 minutes and ethyl gallate (molecular weight = 197) at 22.497 minutes. We confirmed the structure of both compounds using nuclear magnetic resonance and mass spectrometry, and RT was compared with a commercial standard; quantitative analysis reveals that free gallic acid in our fraction is less than 5% dry matter base of the total fraction (4.88% calculated by external standard method) and free ethyl gallate is 1.84% of the dry matter base.

Additionally, we have previously shown that P2Et induces K562 cell death characterized by mitochondrial membrane potential (Δψm) leakage, caspase 3 activation, DNA fragmentation, and reduced tumor cell clonogenic capacity. Also, pretreatment of K562 cells with the P2Et fraction enhances the cytotoxicity of DX. In the breast cancer cell line MCF7, the P2Et fraction exhibited adjuvant activities when combined with all the tested drugs, suggesting a specific mechanism related to the biological characteristics of cell lines. ²¹ Recently, we demonstrated that the P2Et fraction was also active in 4T1 cells via the same mitochondrial-dependent mechanisms, delaying in vivo primary tumor growth and metastasis. ²² Several mechanisms can be implicated in this antitumor activity. In this study, we hypothesized that P2Et may induce the activation of the IR, and indeed, we found that the vaccination of mice with t-P2Et induces an IR concomitant with delayed tumor growth in vivo.

Apoptosis has been considered a silent cell death process that does not activate an IR; however, Kroemer and colleagues recently showed IR activation induced by apoptotic cells expressing CRT with the late secretion of HMGB-1 and ATP, with the subsequent activation of dendritic cells (DC). ⁸ The P2Et fraction induces apoptosis of the 4T1 tumor cell line as previously observed but concomitant with CRT expression on the plasma membrane and the mobilization of HMGB-1 from the nuclei to cytoplasm. This mechanism is similar to that of DX, which is known to induce ICD. ⁴⁷ Our results support the few studies in this field that have explored the possibility that complex fractions derived from plants inducing apoptotic cell death ²² could facilitate the induction of an IR. ¹⁸

Cytotoxic antineoplastic agents mediate their therapeutic effects in a cancer cell–autonomous fashion; however, some chemotherapeutics also indirectly inhibit tumor growth via the immune system.^48,49 Cells resistant to chemotherapy become more sensitive to killing by cytotoxic TL by a mechanism of immunogenic modulation, ⁵⁰ opening the way to establish a compromise between nontoxic chemotherapy and immune activation to improve survival of cancer patients. ⁴⁷ The role of herbal medicine in the induction of IR has not been studied in detail until now and could become an important component in the adjuvant treatment of cancer.

The induction of an IR as a consequence of chemotherapy involves several factors, including a local inflammatory environment as well as the chemotaxis of APC capable of being activated in situ to perform effective cross-priming and generation of a protective IR. One of the in vivo models to study this phenomenon is vaccination with dying tumor cells, which is only protective if it is capable of activating APC, ⁵¹ and subsequently CD4⁺ TL activation, which participates in the sensitization of CD8⁺ T cells ⁵² that, ultimately, target tumor cells resistant to chemotherapy.

In this study, we provide evidence that the vaccination of mice with t-P2Et promotes the early recruitment of CD3⁺ TLs in PLNs, replaced by a mixed CD3⁺/CD3⁻ cell population that might also be involved in tumor growth inhibition.^53,54 The consequence of CD3⁺ T cell mobilization is undoubtedly the activation and generation of cytokine-producing T cells, as we observed in this model. In fact, the ability of a single cell to produce several cytokines simultaneously is referred to as multifunctionality, which translates to a greater protective capacity of the immune system. ¹² In our work, we observed that vaccination with t-P2Et induces a higher frequency of multifunctional CD4 and CD8 TL than t-DX, commonly known as the best inducer of ICD. Interestingly, CD8 TL can only be induced if a cross-priming mechanism occurs, so effective antigen processing is also induced by this therapy; however, it will be necessary to evaluate the long tenure of memory T cells generated.

As mentioned above, not only the cells producing Th1 cytokines, such as IFN-γ, but also the cells producing Th2 cytokines, such as IL-4 or IL-5, are important in the antitumor IR. Th2 cells have been involved in long-term protection in cancer patients,^27,55 possibly by the activation of other responder populations, including NK cells, eosinophils, or B cells^56,57 or the modulation of the inflammatory response. In fact, in our model, we further observed that this t-P2Et vaccination induces the generation and activation of TLs producing IL-4 and IL-5 after stimulation with 4T1 lysate; this response was maintained even 26 days after vaccination and orthotopic tumor transplantation. Dobrzanski et al, ⁵⁸ using a B16-OVA melanoma model, demonstrated that tumor-reactive effector Tc2 cells accumulate at the tumor site and induce regression at an early stage of tumor development, independent of perforin but dependent on IL-4 and IL-5. Moreover, in surviving cancer patients, both Th1 and Th2 clones producing IFN-γ, TNF-α, IL-5, and IL-13 have been shown to exist. ⁵⁹ It is possible that IL-4 contributes in vivo to the stimulation of the dependent cytotoxic activity of perforin and granzyme B in NK cells ⁵⁵ or to the induction of Dendritic cells (DC) maturation able to activate NK cells. ⁶⁰ Other populations can then be selectively recruited after Th2 signals. ⁵⁴

The induction of a multifunctional cytokine profile containing Th2 cytokines and particularly IL-5 after t-P2Et vaccination together with the reduction of tumor size in vivo is surprising. It has recently been reported that IL-5 secretion after a second antigenic stimulus might reflect a strong and durable signal that allows cells to reach their final stage of differentiation. ⁶¹ The generation of these cells together with the multifunctional Th1 T cells and their role in protection against subsequent tumor metastasis should be evaluated together with the mechanisms involved in the delay of tumor cell growth.

Animal models cannot precisely reproduce what will happen in patients; however, the beneficial role of natural polyphenols on human health and specifically in cancer have been widely documented, ⁶² which reinforces the idea that the results observed in this study may be relevant for cancer patients. We will evaluate the effect of P2Et treatment on tumor growth in immunodeficient NOD mice to investigate the actual role of IR in tumor growth control. Additionally, we are evaluating the role of principal compounds present in the P2Et fraction in the control of tumor growth and ICD induction in order to identify the synergistic or antagonistic mechanisms present in the fraction; however, we believe that the entire fraction will be used in cancer treatment.

In summary, the induction of an IR by a natural fraction obtained from a folk plant used for cancer treatment can help explain the in vivo antitumor activity of this and other herbal preparations. These data, combined with the low toxicity of such compounds, emphasize the need for further development of therapies that include complex natural products.